Article Text

SAT0234 Confirmed High Prevalence of Axial Spondyloarthritis in a Validation Study of Primary Care Patients with Chronic Low Back Pain (Cafaspa-2)
  1. L. Van Hoeven1,2,
  2. J. Luime2,
  3. M. Hazes2,
  4. M. de Buck3,
  5. A. Weel1
  1. 1Rheumatology, Maasstad Hospital
  2. 2Rheumatology, Erasmus Medical Centre, Rotterdam
  3. 3Rheumatology, MC Haaglanden, Den Haag, Netherlands


Background In 2010 we presented a prevalence of 21% fulfilling the ASAS classification criteria for axial spondyloarthritis (aSpA) in primary care patients with chronic low back pain (CLBP) (CaFaSpA 1 study)1. In addition, a referral model was developed to help general practitioners (GP) to pick up patients at risk for aSpA. If valid, these data might have a considerable impact for GPs, since CLBP is one of the most frequent musculoskeletal complaints in primary care. We therefore replicated the study in different primary care practices to assess whether our initial findings were valid.

Objectives First to estimate the prevalence of aSpA in primary care patients with CLBP classified by the ASAS criteria. Secondly to externally validate the CaFaSpA 1 referral model which was develop for GPs to identify CLBP patients at risk for aSpA.

Methods We conducted the CaFaSpA 2 study; a stratified, cross-sectional study in primary care patients with CLBP, aged 18-45 years. Patients were identified from GP records by the ICPC code L03 and recruited by two strata of symptom duration. Assessments included inflammatory back pain (IBP) questionnaires (ASAS, Calin and Berlin), a standardized history and physical examination, HLA-B27 and CRP. Conventional radiography and MRI of the sacroiliac joints (SIJ) were scored by one out of two experienced MSD radiologist without having any clinical information. Sacroiliitis was defined by mNY criteria and MRI evaluation followed ASAS recommendations2. Definite aSpA was defined by the ASAS criteria of aSpA2. Descriptive statistics, logistic regression and validation methods in R (version 2.15.2) were used to validate the CaFaSpA 1 referral model.

Results 579 patients participated and were stratified in two groups, symptom duration shorter than 5 years (n= 270, 42% male, age 33.8 (sd 7.4)) and longer than 5 years (n=309, 40% male, age 37.6 (sd 6.1). The overall point prevalence was 17.3% (n=100). No difference was seen related to symptom duration: 17.0% (< 5 yrs) vs 17.5%(≥5 yrs).

The CaFaSpA 1 referral model consists of three variables, a positive ASAS IBP questionnaire, a positive family history of aSpA and a good response to NSAIDs (AUC 0.74, sensitivity 0.70, specificity 0.65, at the cut of 2.28). The performance of the model declined in CaFaSpA 2 with an AUC of 0.67, se of 0.60 and sp of 0.62. No difference was seen between the two groups of symptom duration.

Conclusions In the CaFaSpA 2 validation study the high prevalence of aSpA was confirmed among primary care patients with CLBP. The CAFASPA-1 referral model performed moderately in CAFASPA-2 and analyses for updating with other variables to be truly discriminative in primary care are under construction.


  1. Hoeven van L, Luime J, Han H, Weel AEAM. Striking prevalence of axial spondyloarthritis in primary care patients with chronic low back pain; a cross sectional study. Arthritis and rheumatism 2010;62:2180.

  2. Sieper J, Rudwaleit M, Baraliakos X, et al. The Assessment of SpondyloArthritis international Society (ASAS) handbook: a guide to assess spondyloarthritis. Annals of the rheumatic diseases 2009;68 Suppl 2:ii1-44.

Disclosure of Interest L. Van Hoeven: None Declared, J. Luime: None Declared, M. Hazes: None Declared, M. de Buck: None Declared, A. Weel Grant/research support from: Unrestricted grand from Abbott

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.